R ESEARCH ARTICLE ScienceAsia 34 (2008): 163–168 doi: 10.2306/scienceasia1513-1874.2008.34.163 Treatment of effluent from shrimp farms using watermeal ( arrhiza) Tawadchai Suppadita*, Wisakha Phoochindaa, Saowanit Phutthilerphongb, and Chalida Nieobubpac a School of Social and Environmental Development, National Institute of Development Administration, Bangkok, 10240, Thailand. b TOA Paint (Thailand) Co., Ltd., Bangsaothong, King Amphur Bangsaothong, Samuthprakarn 10540, Thailand. c TEAM Consulting Engineering and Management Co., Ltd., Klong Kum, Bueng Kum, Bangkok 10230, Thailand. * Corresponding author, E-mail: [email protected], [email protected] Received 17 Oct 2007 Accepted 2 Jan 2008

ABSTRACT: Watermeal (Wollfia arrhiza) is a small aquatic which we have used to treat water from black tiger shrimp ponds. The relationship between watermeal biomass and treatment length, the changes in water quality parameters, and N-balance were evaluated for the treatment of black tiger shrimp farm effluent in low-salinity areas. A biomass of 12 g of watermeal per litre of shrimp farm effluent and a treatment period of 30 days were found to provide the best conditions for the growth of watermeal, and the quality of the treated effluent in terms of biological oxygen demands, suspended solids, total phosphorus, nitrate, total ammonia nitrogen, and total Kjeldahl nitrogen. The pH and salinity were similar for each level of biomass. The watermeal biomasses of 4–12 g per litre of effluent were suitable for watermeal survival over time. Since watermeal can fix nitrogen from the atmosphere, it can grow very well in effluent containing a low level of nitrogen, maintaining the N-balance.

KEYWORDS: biomass, , shrimp farm, water plant, water quality

INTRODUCTION cause many environmental problems. These environmental problems include poor water quality Penaeus monodon or black tiger shrimp is and quantity, as well as a variety of pollutants in the an economic species cultured in several tropical water that are the result of substances used at each countries, including Thailand, yielding a high annual stage of shrimp production, preparation and cleaning revenue. Because of this, P. monodon culture has been of the ponds, rearing and harvesting of shrimp6. expanded from along the shoreline to brackish, and In each of these steps, the water can be contaminated then to low-salinity areas. In 1985, the area devoted with organic substances, discarded feed, soil sediment, to P. monodon culture in Thailand was about 408 km2 and plankton. Wastewater from shrimp ponds is and had increased to 744 km2 by 20041,2. Raising released into canals on the farm and/or public water P. monodon in low-salinity areas has been increasing sources where aquatic are found in abundance5. because this species of shrimp belongs to the It is possible that the aquatic plants could absorb Euryhaline group, which can survive well in variable organic substances from the wastewater allowing salinity ranging from 5–30 ppt, or even in almost pure farmers to recycle the water. Raising P. monodon in fresh water3. Normally, a salinity of 0–10 ppt hinders a water-recirculatory system is a way to lower the the growth of disease-causing bacteria that infect cost of production, lower the risk of infections, and freshwater and seawater animals. Accordingly, raising possibly increase the farmers’ income. P. monodon in low-salinity water can reduce the risk For all of these reasons, the idea of studying the of infection by Vibrio harveyi4, especially the type treatment of effluent fromP. monodon ponds by using with luminescent characteristics5. watermeal (Wolffia arrhiza (L.) Wimm) emerged. The As a result of the expansion of P. monodon culture plant grows in natural water sources. Apart from rapid into low-salinity areas, developments have been growth and reproduction, watermeal provides high made in husbandry methods. The traditional method, nutrition, especially protein, so it can be used to make the natural one, has given way to modern techniques animal feed, and is possibly even suitable for human which, without a proper management system, can consumption.

www.scienceasia.org 164 ScienceAsia 34 (2008)

MATERIALS AND METHODS from December 2005 to April 2006. Effluent totalling 1,100 l was randomly collected after the shrimp harvest. The quality of the effluent, the biomass of The water quality was analysed at the Faculty of the watermeal, the nitrogen-balance (N-balance) of Public Health, Mahidol University, Bangkok prior to the effluent, and the watermeal were assessed ina the start of the experiment. Watermeal was obtained 5 x 3 factorial arrangement with 4 replications in a from local markets located at Muang District, completely randomized design, with the first factor Chaiyaphoom Province. This was washed, planted in being the watermeal biomasses of 0, 0.2, 0.4, 0.6, and clean water for 7 days, sampled randomly, and analysed 0.8 kg/50 l of effluent and the second factor being the for total Kjeldahl nitrogen (TKN) prior to the start of experimental time period of 10, 20, and 30 days. The the experiment. data were subjected to an analysis of variance and the The shrimp pond effluent was placed into comparisons among means were made using Duncan’s 20 black plastic containers (50 l/container). Varying new multiple range test of the Statistical Analysis weights of watermeal (0, 0.2, 0.4, 0.6, and 0.8 kg) System (SAS version 6.12)7. were placed in the containers for each treatment The ponds were located at the Pichet Farm, (4 replications) and reared in an experimental house Bansang District, Prachinburi Province. A closed for 30 days. zero-water exchange system was used for the ponds in Effluent from each treatment was sampled at the low-salinity area. The experiment was carried out 10.00 a.m. at the start of the experiment and after

70 120 * * * * * * * * * * 60 100

50 0 kg/50 L 80 0 kg/50 L 40 0.2 kg/50 L 0.2 kg/50 L 0.4 kg/50 L 60 0.4 kg/50 L 30 0.6 kg/50 L SS (mg/L) 0.6 kg/50 L BOD (mg/L) 40 0.8 kg/50 L 20 0.8 kg/50 L

10 20

0 0 0 10 20 30 0 10 20 30 Days after Rx Days after Rx a b

1.0 * * * * * 0.8 * * * * * 0.9 0.7 0.8 0.6 0.7 0 kg/50 L 0 kg/50 L 0.5 0.6 0.2 kg/50 L 0.2 kg/50 L 0.5 0.4 kg/50 L 0.4 0.4 kg/50 L

TP (mg/L) 0.4 0.6 kg/50 L 0.6 kg/50 L

Nitrate (mg/L) 0.3 0.3 0.8 kg/50 L 0.8 kg/50 L 0.2 0.2 0.1 0.1 0.0 0.0 0 10 20 30 0 10 20 30 Days after Rx Days after Rx c d

7 35

6 30 * * * * *

5 25 0 kg/50 L 0 kg/50 L

4 0.2 kg/50 L 20 0.2 kg/50 L 0.4 kg/50 L 0.4 kg/50 L 3 15

0.6 kg/50 L TKN (mg/L) 0.6 kg/50 L TAN (mg/L) * * * * * 0.8 kg/50 L 2 0.8 kg/50 L 10

1 5

0 0 0 10 20 30 0 10 20 30 Days after Rx Days after Rx e f – Fig. 1 BOD (a), SS (b), TP (c), NO3 (d), TAN (e) and TKN (f) at different watermeal biomasses and periods of treatment. * indicates p<0.05 vs other periods of treatment.

www.scienceasia.org ScienceAsia 34 (2008) 165

10, 20 and 30 days. Water quality was analysed for 0.8 kg/50 l, the SS increased. The reasons for this biological oxygen demand (BOD), suspended solids were the same as for the BOD. (SS), pH, salinity, total phosphorus (TP), nitrate – (NO3 ), total ammonia nitrogen (TAN), and TKN using pH standard methods8. In each replication, effluent was There was no significant difference (p>0.05) kept in one litre plastic bottles, cooled in an icebox, between the various watermeal biomass treatments and sent to the laboratory. At the end of the experiment, and the length of treatment time. It was also found that watermeal in each replication was filtered using a net in experimental units with 0.2–0.8 kg of watermeal, for about 5 min. The biomasses were obtained after the watermeal was still adapting to its environment and the effluent was sampled and analysed for TKN using its growth may have been in a lag phase. Therefore, the AOAC method9. its was incomplete and some of the

The N-balance was studied from the total watermeal died, resulting in an accumulation of CO2, – 2– 2– 11 nitrogen in the system containing the TKN value NO3 , SO4 , PO4 and TAN . Although TAN caused in the effluent and the watermeal (before and after an increase in pH12, other compounds suspended the experiment), TKN losses due to volatilization, in the treated effluent caused it to decrease. As a transformation and collected water samples, and consequence, the treated effluent was in a slightly basic N fixation from the atmosphere by the watermeal. state.

RESULTS AND DISCUSSION Salinity As the biomass and treatment period increased, Biological oxygen demand the salinity of the treated effluent increased to 5.2–5.9 ppt The values of BOD at the starting biomasses from an initial salinity of 5.1 ppt (p<0.05). The of the watermeal when different treatment times were maximum salinity was 5.9 ppt when a watermeal applied were different (p<0.05) (Fig. 1a). The water- biomass of 0.6 or 0.8 kg/50 l and treatment time of meal biomass of 0.6 kg/50 l and the treatment time of 30 days were used. This was likely partly due to water 30 days gave the highest BOD removal; the residual evaporation and some water being used for the growth BOD was 19 mg/l. The BOD of the effluent decreased of the watermeal. over time in the 0 to 0.6 kg/50 l treatments. This might be because the watermeal adsorbed organic substances Total phosphorus from the effluent for its growth resulting in a decreasing A comparison between various starting demand for the digestion of organic substances10. biomasses of watermeal and various treatment times However, at the watermeal biomass level of showed differences in the TP of each biomass level 0.8 kg/50 l, the BOD dropped by 34.5% after 10 days. and treatment time (p<0.05) (Fig. 1c). The lowest At days 20 and 30 it was 21.7% and 10% less than reduction in TP was from an initial TP level of 0.91 mg/l the day 0 value for this treatment, respectively. At this dropping to 0.79 mg/l when the watermeal biomass of level, the watermeal died, and most likely led to the 0.8 kg/50 l and treatment time of 30 days were applied, high level of organic substances in the water. whereas the highest reduction of TP left only 0.02 mg/l of TP remaining when a watermeal biomass treatment Suspended solids of 0.6 kg/50 l and a treatment time of 30 days were The levels of SS in the various watermeal used. This might be because the watermeal used biomass treatments were different when different phosphorus in the form of orthophosphate for its treatment times were applied (p<0.05) (Fig. 1b). growth13,14. This result was similar to that of the The highest reduction in SS was obtained when BOD. a watermeal biomass of 0.6 kg/50 l and a treatment time of 30 days were employed. This resulted Nitrate in an SS of 39 mg/l from an initial SS of 102 mg/l. When different watermeal biomasses and – Naturally, even without the watermeal, SS was treatment times were used, the NO3 levels were also decreasing over time. However, with the presence different (p<0.05) (Fig. 1d). After the experiments – of the plant, SS decreased at a faster rate. The SS were completed, the remaining NO3 levels ranged in the form of organic matter might be decomposed from 0.015–0.39 mg/l, which was less than the initial 6 – – by microorganisms and turned into nutrients NO3 value of 0.74 mg/l. The highest reduction of NO3 for the watermeal. As with the findings for was found for the watermeal biomass treatment of 0.6 the BOD levels, in the watermeal treatment of kg/50 l and the treatment time of 30 days, leaving

www.scienceasia.org 166 ScienceAsia 34 (2008)

– a remaining NO3 level of 0.015 mg/l, whereas the – 0.20 lowest reduction of NO3 was found for the watermeal biomass treatment of 0 kg/50 l and the treatment time of 0.15 – 10 days, leaving a remaining NO3 level of 0.39 mg/l. 0.10 * This might be because the watermeal absorbed NO – * 0.2 kg/50 L 3 0.05 for its growth14,15. However, for the treatments with * * 0.4 kg/50 L 0.6 kg/50 L a watermeal biomass of 0.8 kg/50 l, the NO – value 0.00 3 10 20 30 0.8 kg/50 L increased because the watermeal biomass was too -0.05 dense and some of the plants died. The watermeal -0.10 was decomposed by bacteria, which in turn resulted An increase of watermeal biomass (kg) -0.15 – in nitrification, adding NO3 to the effluent. Days after Rx

Total ammonia nitrogen Fig. 2 Watermeal’s biomass at different starting biomasses When different starting biomasses of watermeal and periods of treatment. and treatment times were applied, the concentrations * indicates p<0.05 vs other periods of treatment. of TAN were different (p<0.05) (Fig. 1e). When the watermeal biomass treatment of 0.6 kg/50 l and a when a watermeal biomass of 0.8 kg/50 l was treatment time of 30 days were used, the lowest TAN, used, TAN tended to increase as the treatment time 0.11 mg/l, was achieved. From the initial concentration increased, by up to 5.79 mg/l. This might be because of TAN, which was 2.24 mg/l, the TAN tended to of the death of some watermeal plants. decrease as the biomass and treatment time increased. This might be because the higher the biomass, the higher Total Kjeldahl nitrogen the adsorption of TAN from the water15. However, When different starting biomasses of watermeal

215 mg 42.5 mg 1,400 mg 4 1,400 mg 4 1 1 0 mg 88,143 mg 5 5 0 mg 20,780 mg 2 0 kg/50 l 2 0.2 kg/50 Ll 1,142 mg 1,142 mg 0 mg 6 67,184 mg 6 3 a 43.0 mg 3 b 36.5 mg 7 7

30.0 mg 27.7 mg 1,400 mg 4 1,400 mg 4 1 1 120,194 mg 146,343 mg 5 5 41,560 mg 62,340 mg 2 0.4 kg/50 l 2 0.6 kg/50 l 1,142 mg 1,142 mg 78,437 mg 6 83,800 mg 6 3 c 31.0 mg 3 d 27.3 mg 7 7

702 mg KN in effluent KN in effluent 1,400 mg 4 1 4 1 (before experiment) (after experiment) 137,310 mg 5 KN in watermeal KN in watermeal 83,120 mg 0.8 kg/50 l 2 5 2 (before experiment) (after experiment) 1,142 mg 54,691 mg 6 N fixation from KN loss in volatilization 3 6 3 e 57.0 mg atmosphere and transformation 7 KN loss in collected 7 water samples Fig. 3 N-balance at 0 (a), 0.2 (b), 0.4 (c), 0.6 (d), and 0.8 (e) kg biomasses of watermeal.

www.scienceasia.org ScienceAsia 34 (2008) 167 and treatment times were applied, the TKN levels in biomass treatment of 0.8 kg/50 l was used because the effluent were also different (p<0.05) (Fig. 1f). The some of the plants died due to an excessive density lowest level of TKN contamination in the effluent was of biomass. 0.9 mg/l and was obtained from the biomass treatment of 0.6 and a treatment time of 30 days. This might be CONCLUSIONS because the watermeal used N in the form of TAN – 16 and NO3 for its growth , and this resulted in TKN The biomass level of 0.6 kg watermeal/50 l effluent levels of 0.9–24.6 mg/l. The initial value of TKN was and the treatment time of 30 days showed the best 28 mg/l. In contrast, when the watermeal biomass of performance in water quality improvement in this 0.8 kg/50 l and the treatment times of 20 and 30 days experiment. The growth of the watermeal declined were used, the level of TKN increased, since there was when a biomass concentration of 0.8 kg/50 l was an excess of watermeal that led to its decomposition, employed, because of an excess of the watermeal. transforming it into N in the effluent. The reason for N could be fixed from the atmosphere by watermeal this was similar to this phenomenon for TAN and for its growth. Thus, watermeal can grow very well – NO3 levels. in effluent containing a low level of N. The use of watermeal for treating effluent fromP. monodon ponds Biomass is practical and feasible when optimal conditions Significant differences p( <0.05) in watermeal are met and a suitable biomass concentration and time biomass levels arose as the treatment period period is used. lengthened. The initial biomass level of 0.2 kg/50 l and a treatment time of 30 days yielded the highest ACKNOWLEDGEMENTS increase in biomass, about 0.156 kg, whereas the initial biomass level of 0.8 kg/50 l and a treatment time of The authors would like to thank Pichet Farm for 30 days showed the greatest decrease in biomass, material support. We also highly appreciate the Faculty about -0.086 kg (Fig. 2). This might be because there of Public Health, Mahidol University for permitting was a high density of biomass. It was found that the us to conduct the laboratory study. lower the initial amount of biomass, the higher the survival of watermeal, and the higher the initial amount REFERENCES of biomass, the lower the survival of watermeal. 1. Hemmun P (2000) Study on the variation of N-balance water and soil qualities, phytoplankton in black The total N in the system was composed of the tiger shrimp (Penaeus monodon Fabricius) total N in the effluent before the experiment, which pond in freshwater area at Ratchaburi Province. was equal for all experimental units, and the total M.Sc. thesis, Kasetsart University, Bangkok, N in the watermeal before the experiment, which Thailand. depended on the initial amount of watermeal biomass. 2. Department of Pollution Control (2005) Coastal It was found that the greater the amount of watermeal water quality standards, December, 18. biomass, the higher the level of total N (Fig. 3). 3. Musig Y (1987) Biological capacity in fish pond During the experiment, total N was lost when some II. Department of , Faculty of fishery, effluent was removed for analysis. Also, total N was Kasetsart University, Bangkok, Thailand. lost due to the volatilization and transformation of N. 4. Chongpeephia T (1999) Relationship between However, there was an increase in the total N from bacterial flora and water quality in low salinity the nitrification process of atmospheric nitrogen. This black tiger Shrimp ponds (Penaeus monodon). process increased the total N as the biomass increased. Coastal Aquaculture Division, Department of As a result of the increases in the total N mentioned, Fisheries, Ministry of Agriculture and the results showed an increase in the total N released Cooperatives, Bangkok, Thailand. from the system, which was accumulated in the 5. Suppadit T, Phoochinda W, Bunsirichai P (2005) biomass. The results also showed that the total N Treatment of effluent from shrimp farms using in the treated effluent at the end of the experiment water mimosa (Neptunia oleracea Lour). decreased as the biomass increased whereas the total J. ISSAAS 11, 20–9. N in the watermeal increased as the biomass increased 6. Bunsirichai P (2004) A feasibility study of except for the biomass level of 0.8 kg/50 l. In this effluent treatment from black tiger shrimp pond case, the total N in the watermeal decreased when the by using water mimosa. M.Sc. thesis, National

www.scienceasia.org 168 ScienceAsia 34 (2008)

Institute of Development Administration, Bangkok,Thailand. 7. SAS Institute (1996) Statistical analysis system. SAS institute Inc., North Carolina, U.S.A. 8. APHA AWWA and WEF (1998) Standard methods for the examination of water and waste- water. American Public Health Association, Washington D.C., U.S.A. 9. AOAC (1992) Official methods of analysis of the association of official analytical chemists. Arlington Inc., Washington D.C., U.S.A. 10. Chaiyakum K, Predalumburt Y (1994) Preventive methods for decreasing the impacts to environment from the development of black tiger shrimp domestication in development Form. Thai- land National Institute of Coastal Aquaculture, Department of Fisheries, Bangkok, Thailand. 11. Jiwyam W (1997) Water quality for aquaculture and analytical methods. Department of Fishery, Faculty of Agriculture, Khon Kaen University, Khon Kaen, Thailand. 12. Panyakapo M (2002) Biological wastewater treatment. Silpakorn University Press, Nakhon- pathom, Thailand. 13. Chowwanklang P (1991) Water quality of fisheries. Physics Center Press, Bangkok, Thailand. 14. Pansawad T (2002) Biological nitrogen and phos- phorus removals. 2nd edn. The Environmental Engineering Association of Thailand, Bangkok, Thailand. 15. Tuntulweat M, Tuntulweat M (2004) Chemistry of water and wastewater. Chulalongkorn Univer- sity Press, Bangkok, Thailand. 16. Mahakhan A (2004) Watermeal: high quality food. J. Sci. Technol. 13, 95–101.

www.scienceasia.org